Infercia Voice Diarization

Upload audio/video. The API uploads the original file to Azure Blob Storage, queues diarization, and the worker saves JSON to gender-results/{taskId}.json.

Task ID Force rerun detection

Number of speakers (optional, improves accuracy a lot when known)

Upload audio/video file

Analyze existing blob name

Enroll a known speaker (voiceprint): name + a clean 10-30s sample of only their voice

Result

Waiting...